Whistled Platform Upgraded With Word Recognition

An anonymous reader writes "A few weeks ago, Slashdot featured a cheap platform performing 80FFTs per second to recognize whistles. The platform is open hardware/open source and is aimed for sound processing projects. To this goal, the creator (limpkin) just implemented a simple proof of concept algorithm that will control your lighting once the platform listens to a particular word. A small video has been made to explain the basic concepts of sound recognition to encourage hobbyist to make their own."

Comments

Not A Criticism, But... So What?

[Jane Q. Public]

In the 1970s, a book was published entitled "How To Build Your Own Computer-Controlled Robot". In fact this was one of the books that first got me interested in computing.

The author of the book, a high-school student, built (with the help of his engineer father, I don't want to downplay that) a small robot that had obstacle sensors, light sensors, and some basic mapping capability so it could find its own charging station (not so different from a Roomba today).

But my main point is: it also had effective voice recognition, for simple commands. And the implementation was pretty simple: the audio input was amplified, then sent through 3 notch filters to separate high, medium and low audio frequencies. Each of the 3 frequencies was digitally sampled at about 40kHz. Repeated samples were averaged and saved in a table in memory.

The CPU (and remember, this was a 1970s-era CPU, if I recall an 8080a or a Z-80 or similar) constantly sampled incoming sounds, and when one sufficiently matched one of the stored templates it meant "command received".

It was a simple scheme, and it worked fine. I don't want to detract from this inventor, but in essence he is doing a similar thing. Except instead of using notch filters, he's using FFTs to do the frequency analysis and build (and then compare to) the templates. The ideas aren't all that different.

But personally, I think I'd prefer the old method, as it demonstrably worked at least as well as this, used only a few $ in hardware in addition to the CPU, and was pretty definitely less compute-intensive to achieve.

Keep in mind: that was 40 years ago. Maybe this newer approach has more potential; I don't know. But it certainly doesn't look much different at this time.

Re: Not A Criticism, But... So What?

[Anonymous Coward]

If you can't see why this has more potential, you know nothing about DSP. Yes, for voice, 3 well-chosen fixed bands is mostly enough. But for whistling (as per the original application) or various other sounds, those same three bands will be pretty crap.

Now we could make the 3 analog notch filters tunable via a DAC output, and get good results for one group (at a time) of a wide range of sounds, but that complicates it substantially. Yes, doing it with FFTs uses a lot more computational power -- but when the cheapest microcontrollers commonly used by hobbyists these days are a lot more powerful than the venerable Z80 (FYI, I built a single-board CP/M box back in the day), why in the world shouldn't we use some of that muscle to do more things, and some of it to do the same things better?

Oblig. Demolition Man

[BlueStrat]

Simon Phoenix, after reprogramming Dr. Cocteau's house lights;

"Nah, I changed that."

"Illuminate"

"De-luminate"

"Ah, Isn't that much better?"

(I swear, the US is looking and feeling more and more like the fictional "Greater SanAngeles" from the movie with every day that passes.)

Strat

just imagine...

[Connie_Lingus]

..if he could get it recognize, say. the sound of two handclaps...wouldnt that be something?

Even better...

[3nails4aFalseProphet]

...the sound of one hand clapping.

Prometheus

[TheSkepticalOptimist]

The moment I have to blow anything to use a device I will retire and become a Luddite.